Cell-permeable peptides can penetrate the cell membrane and become internalized either alone or coupled to other molecules. Their value has been recognized especially in vaccination and gene therapy studies (for a review, see Reference 1). Gratton et al. reported recently about the use of these peptides in enhancement of virus-mediated gene delivery in vitro and in vivo (2). They showed that polybasic peptides derived from Drosophila Antennapedia homeodomain (Antp) type 1 (HIV-1) transactivator protein (TAT) improved adenoviral and retroviral transduction in cultured monkey COS-7 cells, bovine aortic endothelial cells, and human umbilical vascular endothelial cells, as well as in vivo when precomplexed with viral vector particles. Based on their results, Gratton et al. (2) suggested that highly positively charged peptides can enhance the transduction by concentrating viral particles to the cell surface and by improving receptor-dependent uptake mechanisms.

Insufficient transduction efficiency is still considered the major problem in gene therapy research. Because high gene transfer rate is particularly important in most cancer gene therapy approaches and Gratton et al. (2) did not test their concept in human tumor cells, we conducted a series of experiments to verify the utility of cell-permeable peptide-complexed virus vectors in cancer gene therapy. In addition to the Antp peptide (RQIKIWFQNRRMKWKK), we used two versions of the HIV-1 TAT peptide: TAT1 (YGRKKRRQRRR, used in most earlier studies) (1) and TAT2 (GRKKRRQRRRPPQ, presumably used by Gratton et al.) (2). Furthermore, two polycationic compounds, polybrene (hexadimethrine bromide) and protamine sulfate, were used to identify the contribution of a plain electrostatic effect (i.e., the reduction of the electrostatic repulsion between negatively charged viral particles and cell membranes). Polybrene has been known to enhance retroviral infection since the late 1960s (3). It is nowadays a commonly used enhancer of retroviral transduction, and its mechanism of action (4,5) and positive effect to adenoviral gene transfer (6) have been elucidated. Protamine sulfate has been designated as a more clinically relevant alternative for polybrene in retroviral gene therapy (7), and its utility in adenoviral gene transfer has been acknowledged (8).

The cell-penetrating peptides were incubated with a serotype 5, E1/E3-deleted adenovirus vector AdTK-GFP (9) and a second generation VSV-G pseudotyped lentivirus vector WOX-TK-GFP (10) as described in the original report (2). Polybrene (8 µg/ mL) and protamine sulfate (5 µg/mL) were added to virus dilutions, and the resulting complexes were then used for transduction of one monkey kidney fibroblast cell line (COS-7) and four different human cancer cell lines representing ovarian carcinoma (SKOV3. ipl, HEY), prostate carcinoma (PC-3), and osteosarcoma (MG-63). The human tumor cell lines were selected due to their known features as targets for viral gene transfer. All of these cell lines were moderate or poor targets for lentiviral and/or adenoviral vectors (9,11), and transduction of these cells would apparently benefit from peptide-mediated enhancement. The results, indicated as proportion of green fluorescent protein (GFP) positive cells, were determined by flow cytometry, and a one-way analysis of variance with Dunnett's post hoc test for multiple comparisons was used for statistical analysis.

To verify the results obtained by Gratton and coworkers (2), we examined the peptide- and polycation-mediated enhancement of viral gene transfer efficiency in COS-7 cells ((Figure 1)). Analysis of TK-GFP positive cells by flow cytometry 2 days (adeno-virus) or 4 days (lentiviras) posttransduction confirmed that the antp peptide can significantly improve adenoviral and lentiviral gene transfer (p <0.001). However, this effect was not as impressive as observed earlier (2) (i.e., From 10% with adenovirus vector alone to almost 95% when complexed with the Antp peptide). In our experiments, the Antp peptide doubled the gene transfer efficiency of both vector types, and a similar effect was obtained with the TAT-derived peptides. Interestingly, polybrene and protamine were able to boost the gene delivery with both vector types equally well (protamine) or even better (polybrene) than any of the peptides. It is difficult to determine, why Gratton and colleagues obtained higher enhancement of adenoviral gene delivery with Antp peptide even though their peptide concentration was similar to that used in our experiment. It is possible that the COS-7 cell populations in two different laboratories may not be completely identical or the quality of the adenovirus and peptide preparations may play a role. Furthermore, there may be minor differences in the complex formation or transduction protocols, resulting in variation in the peptide-mediated enhancement. Taken together, our results with COS-7 cells confirm the enhancement observed by Gratton et al. (2), but also point out that depending on the conditions and materials used, the degree of gene delivery enhancement is likely to vary from laboratory to laboratory.